The main goal of Phase I is to compare the effectiveness of internal sensors versus external sensors for condition monitoring of rotary blood pumps. The fundamental hypothesis of this proposal is that a vibration sensor mounted directly on the blood pump would detect incipient failures with good repeatability, high sensitivity, and specificity. This hypothesis is to be evaluated by using an implanted accelerometer and external microphone to record signals from an animal-implanted vibration generator. This approach will provide information similar to seeded-fault experiments. Various mixes of signal frequencies and strengths representing different failure modes will be introduced using the vibration generator, and signals from the accelerometer and microphone will be analyzed to quantify sensitivity and specificity of sensors as well as the effect of live tissue on signal transmission and damping. This information will not only help in making informed decisions for a Phase II effort for development of a condition monitoring system, but will also be important for development of monitoring systems for similar cardiac prostheses. PROPOSED COMMERCIAL APPLICATION: NOT AVAILABLE